• 한국전자파학회논문지
• /
• 제14권6호
• /
• pp.616-624
• /
• 2003

상수도 누수로 인한 우리나라 수자원의 고갈 때문에 효율적으로 누수를 탐지할 수 있는 방법이 시급하다. 본 논문에서는 지하탐사 레이더(GPR) 기법을 이용하여 누수지점을 탐사하였다. 메탄올이 채워진 아크릴 상자로 누수가 일어난 영역을 구현한 후, 본 실험실에 구축된 GPR 시스템을 사용하여 scale-down 실험을 수행하였다. 본 GPR실험의 타당성은 측정결과가 동일한 상황에서의 FDTD 계산 결과와 거의 일치함을 보임으로써 확인하였다. 누수 분포에 따른 B-scan 영상들을 제시함으로써 GPR 시스템의 누수탐지 가능성을 살펴보았다.

• 한국통신학회논문지
• /
• 제36권10C호
• /
• pp.629-634
• /
• 2011

주차장에서 빈 공간을 분석하는 기술은 주차공간의 효율적인 사용이나 교통이 혼잡한 곳에서 유용하게 사용될 수 있다. 그러나 기존의 주차 공간 분석 방법은 실용적이지 못하거나 빠른 처리속도가 필요하다. 그래서 본 논문에서는 실시간 주차검출에 적합한 주차 모니터링 방법을 제안하고자 한다. 제안된 방법은 그레이레벨 영상을 이용하여 주차여부를 확인하고, 주차공간을 분석하는 방법을 사용하였다. 제안된 방법의 성능을 확인하기 위해 야외 주차장에서 129개의 동영상을 획득하여 실험한 결과 98.5%의 주차 공간 분석에 성공하여 제안된 방법이 주차 공간 분석에 효율적인 것을 입증하였다.

• ETRI Journal
• /
• 제39권3호
• /
• pp.326-335
• /
• 2017

In the uplink transmission of massive (or large-scale) multi-input multi-output (MIMO) systems, large dimensional signal detection and its hardware design are challenging issues owing to the high computational complexity. In this paper, we propose low-complexity hardware architectures of Richardson iterative method-based massive MIMO detectors. We present two types of massive MIMO detectors, directly mapped (type1) and reformulated (type2) Richardson iterative methods. In the proposed Richardson method (type2), the matrix-by-matrix multiplications are reformulated to matrix-vector multiplications, thus reducing the computational complexity from $O(U^2)$ to O(U). Both massive MIMO detectors are implemented using a 65 nm CMOS process and compared in terms of detection performance under different channel conditions (high-mobility and flat fading channels). The hardware implementation results confirm that the proposed type1 Richardson method-based detector demonstrates up to 50% power savings over the proposed type2 detector under a flat fading channel. The type2 detector indicates a 37% power savings compared to the type1 under a high-mobility channel.

• Advances in robotics research
• /
• 제2권1호
• /
• pp.33-44
• /
• 2018

This paper presents a novel feature detection algorithm called depth-hybrid speeded-up robust features (DH-SURF) augmented by depth information in the speeded-up robust features (SURF) algorithm. In the keypoint detection part of classical SURF, the standard deviation of the Gaussian kernel is varied for its scale-invariance property, resulting in increased computational complexity. We propose a keypoint detection method with less variation of the standard deviation by using depth data from a red-green-blue depth (RGB-D) sensor. Our approach maintains a scale-invariance property while reducing computation time. An RGB-D simultaneous localization and mapping (SLAM) system uses a feature extraction method and depth data concurrently; thus, the system is well-suited for showing the performance of the DH-SURF method. DH-SURF was implemented on a central processing unit (CPU) and a graphics processing unit (GPU), respectively, and was validated through the real-time RGB-D SLAM.

• 전기전자학회논문지
• /
• 제23권2호
• /
• pp.448-453
• /
• 2019

모바일 앱의 신뢰성 향상과 개발 환경 변화라는 제약 조건을 모두 만족시키려면 모바일 앱의 동작을 자동으로 검증할 필요가 있다. 모바일 앱의 동작 검증 과정에서 다양한 이슈가 발생하나, 사용 상태 변화 탐지도 중요한 이슈 중 하나이다. 본 논문에서는 모바일 앱의 사용 상태 변화 탐지를 위하여 딥뉴럴넷을 이용하여 모바일 앱 GUI의 UI 위젯을 인식한 후 인식된 위젯간의 관계를 그래프로 변환하고, 변환된 그래프의 그래프 엔트로피를 계산하여 사용 상태 변화를 감지하는 방법을 제안한다. 제안 방법은 SIFT(Scale-Invariant Feature Transform)에 기반한 감지 방법과 비교되었으며 20개의 실제 모바일 앱의 동작 데이터를 통해 검증한 결과 대부분의 경우 제안 방법이 우수하나, 엔트로피 계산이 어려울 때는 제안 방법의 성능이 저하됨을 확인하였다.

• Nuclear Engineering and Technology
• /
• 제51권6호
• /
• pp.1616-1625
• /
• 2019

Measuring occurrence times of random events, aimed to determine the statistical properties of the governing stochastic process, is a basic topic in science and engineering, and has been the subject of numerous mathematical modeling approaches. Often, true statistical properties deviate from measured properties due to the so called dead time phenomenon, where for a certain time period following detection, the detection system is not operational. Understanding the dead time effect is especially important in radiation measurements, often characterized by high count rates and a non-reducible detector dead time (originating in the physics of particle detection). The effect of dead time can be interpreted as a suitable rarefied sequence of the original time sequence. This paper provides a limit theorem for a high rate (diffusion-scale) counter with extendable (Type II) dead time, where the underlying counting process is a renewal process with finite second moment for the inter-event distribution. The results are very general, in the sense that they refer to a general inter arrival time and a random dead time with general distribution. Following the theoretical results, we will demonstrate the applicability of the results in three applications: serially connected components, multiplicity counting and measurements of aerosol spatial distribution.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.153-166
• /
• 2022

Identifying fine cracks in steel bridge facilities is a challenging task of structural health monitoring (SHM). This study proposed an end-to-end crack image segmentation framework based on a one-step Convolutional Neural Network (CNN) for pixel-level object recognition with high accuracy. To particularly address the challenges arising from small object detection in complex background, efforts were made in loss function selection aiming at sample imbalance and module modification in order to improve the generalization ability on complicated images. Specifically, loss functions were compared among alternatives including the Binary Cross Entropy (BCE), Focal, Tversky and Dice loss, with the last three specialized for biased sample distribution. Structural modifications with dilated convolution, Spatial Pyramid Pooling (SPP) and Feature Pyramid Network (FPN) were also performed to form a new backbone termed CrackDet. Models of various loss functions and feature extraction modules were trained on crack images and tested on full-scale images collected on steel box girders. The CNN model incorporated the classic U-Net as its backbone, and Dice loss as its loss function achieved the highest mean Intersection-over-Union (mIoU) of 0.7571 on full-scale pictures. In contrast, the best performance on cropped crack images was achieved by integrating CrackDet with Dice loss at a mIoU of 0.7670.

• International Journal of Advanced Culture Technology
• /
• 제10권2호
• /
• pp.252-259
• /
• 2022

Artificial intelligence is used in fusion with image processing techniques using cameras. Image processing technology is a technology that processes objects in an image received from a camera in real time, and is used in various fields such as security monitoring and medical image analysis. If such image processing reduces the accuracy of recognition, providing incorrect information to medical image analysis, security monitoring, etc. may cause serious problems. Therefore, this paper uses a mixture of YOLOv4-tiny model and image processing algorithm and uses the COCO dataset for learning. The image processing algorithm performs five image processing methods such as normalization, Gaussian distribution, Otsu algorithm, equalization, and gradient operation. For RGB images, three image processing methods are performed: equalization, Gaussian blur, and gamma correction proceed. Among the nine algorithms applied in this paper, the Equalization and Gaussian Blur model showed the highest object detection accuracy of 96%, and the gamma correction (RGB environment) model showed the highest object detection rate of 89% outdoors (daytime). The image binarization model showed the highest object detection rate at 89% outdoors (night).

• BMB Reports
• /
• 제38권2호
• /
• pp.191-197
• /
• 2005

Although the human genome has been nearly completely sequenced, the functions and the roles of the vast majority of the genes, and the influences of single nucleotide polymorphisms (SNPs) in these genes are not entirely known. A modified mutation detection method was developed for large-scale cloning of the possible SNPs between tumor and normal cells for facilitating the identification of genetic factors that associated with cancer formation and progression. The method involves hybridization of restriction enzyme-cut chromosomal DNA, cleavage and modification of the sites of differences by enzymes, and differential cloning of sequence variations with a designed vector. Experimental validations of the presence and location of sequence variations in the isolated clones by PCR and DNA sequencing support the capability of this method in identifying sequence differences between tumor cells and normal cells.

• 제어로봇시스템학회논문지
• /
• 제9권2호
• /
• pp.107-116
• /
• 2003

This paper presents a systematic design methodology for fault tolerant controller against a fault in actuators and sensors of linear stochastic systems with uncertainties. The scheme is based on fault detection and diagnosis(isolation and estimation) using a bank of robust two-stage Kalman filters, and accommodation of the actuator fault by eigenstructure assignment and immediate compensation of the sensor's faulty measurement. In order to clarify the fault feature in test statistics of residual, noise reduction method is given by multi-scale discrete wavelet transform. The effectiveness of our approach Is shown via simulations for a VTOL(vertical take-off and landing) aircraft subjected to parameter variations, external disturbances, process and sensor noises.